Hemodialysis is the primary life-sustaining treatment for patients with end-stage kidney disease, which can result in a range of negative reactions (e.g. depression and anxiety) and positive change (e.g. post-traumatic growth, PTG). To further understand the post-traumatic response patterns of patients who are going through hemodialysis, dividing them into homogeneous subgroups helps to uncover hidden information. This cross-sectional study evaluated 274 patients (172 males and 102 females) undergoing hemodialysis at a tertiary hospital between October and November 2022. Latent profile analysis (LPA) was used to identify subgroups of patients based on their PTG, depression, and anxiety. The results identified three profiles: post-traumatic depression (n = 90, 32.9%), general moderate growth (n = 116, 42.3%) and high appreciation-power (n = 68, 24.8%). The classification quality was good (Entropy = 0.86). Univariate and multiple logistic regression analysis were conducted to test potential influencing factors. Furthermore, Back propagation (BP) neural network model was employed to examine the ranking of influencing factors across different profiles. Perceived health control (100.0%), social network (83.4%), meaning in life (70.2%), self-disclosure (64.4%), self-rated health (45.0%), and exercise (31.4%) were identified as differential predictors of profile members (all p < 0.05), listed in order base on their degrees of influence. The findings revealed the heterogeneity of post-traumatic response patterns in hemodialysis patients, categorizing their post-traumatic responses into three distinct patterns. In the future, in the treatment and care of hemodialysis patients, it will be more meaningful to provide targeted interventions for the different characteristics of patients' PTG, depression and anxiety.